The System Behind Self-Driving: Waymo’s Dmitri Dolgov

Waymo is now delivering hundreds of thousands of fully autonomous rides each week — but getting there required more than better models. It meant building a complete system for training, evaluating, and deploying a driver in the real world. In this episode — originally aired on the Cheeky Pint podcast — Waymo Co-CEO Dmitri Dolgov joins John Collison to break down how self-driving actually works today: from sensor fusion across LiDAR, radar, and cameras, to simulation, “critic” models, and the role of AI in decision-making. They also explore why full autonomy is fundamentally different from driver-assist, what it takes to scale globally, and how recent advances in AI are reshaping the path forward.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  When you're driving around, or being driven around, you know, we think about what we're building as a driver. I can imagine building a big model that understands how the physical world works and understands the important properties of what it means to drive, the social aspects of driving, and what it means to be a good driver as opposed to a bad one. I would say that we've clearly moved past the stage of scientific research and kind of deep core technology development to this new phase of accelerated global scaling and deployment.

• 0:40Erik Torenberg?

  Waymo is now doing nearly half a million fully autonomous rides a week across multiple cities, a shift from long-term research to real-world scale. In this episode, originally aired on the Cheeky Pint podcast, Waymo co-CEO Dmitry Dolgov joins John Collison to break down how they built the system behind it. From the sensor stack and why LiDAR still matters, to the role of simulation and critic models in training the AI, they also get into why driver assist won't naturally evolve into full autonomy, what it takes to scale globally, and how the product itself is changing, from custom-built vehicles to entirely new economies of ride hailing.

• 1:23Speaker 02

  Dmitry Dolgov is co-CEO of Waymo. He joined Google's self-driving car project in 2009 as one of its first engineers, and was repeatedly promoted until he took it over in 2021. Waymo is Google's most successful moonshot, and now provides over 500,000 fully autonomous rides each week. Cheers, by the way.

• 1:40Speaker 00

  Yeah, cheers.

• 1:41Speaker 02

  You grew up in Russia, right?

• 1:42Speaker 00

  Yes, I grew up in Russia. Then I was actually Soviet Union.

• 1:46Speaker 02

  Right, right, exactly.

• 1:47Speaker 00

  My dad is a physicist, so...

• 1:51Speaker 00

  The Soviet Union started falling apart and then he got, had a position, a visiting position in university, in Kyoto University for a year. We moved there as a family and then he went to Berkeley and then kind of tagged along. And then I ran out of, I graduated from high school, I was thinking about the next thing I wanted to do and I really liked that, that, that technical school in Russia.

• 2:14Speaker 02

  The Russians are serious about the physics.

• 2:16Speaker 00

  They are, they are. So I went back to Russia and I got my bachelor's and master's.

• 2:20Speaker 02

  What year was it that you went back to Russia?

• 2:22Speaker 00

  Nineteen ninety-four.

• 2:23Speaker 02

  Okay. So that was kind of almost peak Russian optimism in a sense where it was opening up.

• 2:29Speaker 00

  It was, it was. Yeah, yeah. No, I actually remember talking to my mom about it and of course, my parents grew up in the Soviet Union. They've seen it. They were born right before the war and then they saw, you know, they lived through some really tough times. And I remember talking to my mom and saying, she, she, you know, in fact, I got my green card here in the US before I went back and she insisted that I do it. And I was, actually at the time, I wasn't thinking of coming back. But then I was pretty excited about where Russia is and the trajectory it's on. And, you know, being nine, young and naive, I was like, there's no, there's no turning back.

• 3:08Speaker 02

  And so why did you decide to come back?

• 3:09Speaker 00

  Um.

• 3:10Speaker 02

  There's more of a play by play than.

• 3:11Speaker 00

  Yeah, yeah. No, school, it was pretty clear to me. Like, I wanted to continue studying math and computer science. And while the undergrad and masters that I got in physics and applied math, that I think was still an incredibly strong kind of foundational, you know, school of Russian math and science. Graduate school, it was very clear to me that the best way to do it was in the US. So I came back.

• 3:39Speaker 02

  I'm struck by the founders of the two most valuable UK companies are Russian math nerds who both went to the same school, Nikolai Ash Revolut and Alex Girkko, Alex Girkko at XGX. But yeah, it's a, it's a strong diaspora.

• 4:01Speaker 00

  There is a company not far from here where one of the founders also has somewhat pedigree. A company that.

• 4:09Speaker 02

  Exactly. Exactly. You're the classic engineering interview question of, you know, what happens when I type google.com and hit enter? As you know, talk me through, you know, whatever you like, you know, HTTP and DNS and BG. You can go down to whatever level of stack you want. Do you want to maybe just describe when I take a ride in a Waymo today? What's happening at a technical level? Like, what is the architecture?

• 4:34Speaker 00

  Let me answer your question. I was happening in real time, but this is going to be only a part of the story because we're going to be talking about kind of the inference, the real time inference part of it. And if we want to have a deeper, richer technical conversation, I think it would be interesting also to zoom out and talk about the entire ecosystem of what goes into building, evaluating and deploying the Waymo driver. But when you're driving around or being driven around, say, you know, we think about what we're building as a driver. Obviously, it's not a car, so it has a number of sensors that are positioned around the vehicle.

• 5:13Speaker 00

  We use three different sensing modalities. There's cameras, there's lighters or lasers and radars. You know, those are the primary ones. There are also microphones, directional microphone arrays, but those are the primary three for sensing the world. They all have very nicely complementary physical properties. They all have 360 degree coverage around the vehicle, so the Waymo driver sees kind of 360, you know, all the time. So all of the data goes into a computer, you would expect. And the software that process now it's all AI. I can see a specialized AI in the physical world, so it processes the sensor data.

• 5:56Speaker 00

  Nowadays, you know, talk about it in the, you know, using AI terminology as, you know, encoders that take this data in. And then there's the kind of the decoder, the action, you know, the generative part, if you will, in the car. And the generative task there is to, you know, figure out how to drive, right? And that is of course connected through kind of a specialized interface to the car where we can actuate the vehicle. And that's why you see the steering wheel turn and it drives you around.

• 6:25Speaker 02

  Okay, so I get into my car, there's three main. Families of sensors, LiDAR, radar and cameras. And then it is using that to first build a model of what's going on in the world. You know, where are all the other cars and things like that? And then you say, make decisions and then actuate that with the cars. That is the the system that you're living in. And is all that inference done locally? Or presumably, yes, nothing's in the cloud? Nothing real time?

• 6:50Speaker 00

  Nothing real time without. And there are some things that can happen in the cloud, but they're not required.

• 6:56Speaker 02

  Got it. What's an example of a nice to have that happens in the cloud?

• 6:59Speaker 00

  You can imagine a situation where we do. You know, some of it is not directly related to the Tesco driving. But say after you leave the car, we want to check that, you know, the car is not dirty. You didn't leave anything there. If you did leave, you know, an item, well, if you left in a mess, then you want to send the car to one of our depots, get it cleaned up. If you left an item there, your phone, or we want to detect that and then, you know, send it to our lost and found and let you know. Right? So that we do with the kind of a by asking a model that's actually lifts off board as opposed to having to put it on the car, right? Because it's not a real time task related to, you know, the driving.

• 7:44Speaker 00

  So that's one example of something that.

• 7:46Speaker 02

  There are all these debates that go on on Twitter around self-driving. So I can think of, you know, end to end versus the more kind of modular approach. There's cameras only versus array of sensors. And I can't tell, are these debates actually interesting to an expert in the field? Or do you think these are just settled matters and they're just grist for the algorithm?

• 8:13Speaker 00

  I understand where the questions are coming from. I do find that kind of often the way they're posed and the way the debate happens is losing a lot of the nuance and a lot of.

• 8:29Speaker 00

  Detail that really matters are, to me, the most interesting technical questions are in that level. Because the way we think about the building the Waymo driver, it starts with a large onboard foundation model. I can imagine building a big model that understands how the physical world works and understands the important properties of what it means to drive, the social aspects of driving, and what it means to be a good driver as opposed to a bad one. So that's the foundation. Then we specialize it into, what we call it, three main onboard teachers.

• 9:13Speaker 00

  There are still large, high-capacity onboard models. There's the Waymo driver, there is the simulator, and then there is the critic. And those then get distilled into smaller models that you can run inference on faster. So the Waymo driver becomes the backbone, the backbone of what's in the car. The simulator, of course, is what powers our synthetic generative environment that can run on the cloud for training and for evaluation in close of the system. And the critic, they value it. Does the simulator ever run locally? No. No, it doesn't. Yeah. However, what I think is interesting, in a way, the way the decoder works, the way the model works, if you think about the generative task in the simulator,

• 9:58Speaker 00

  of creating those realistic worlds and how other people behave, how cars, pedestrians, cyclists, in order... And the task that you have to solve on the car in real time, there is this fundamental shared capability of understanding how these objects relate to each other and predicting what they might do in the future if you are running on the car, and then generating some sampling of those probabilistic behaviors in the simulator. So it's a different model, but there is, this is why the shared foundation model is able to power both. And similarly, if you think about the critic, like the job of the critic is to find interesting events and then be opinionated

• 10:43Speaker 00

  about what's good behavior and what's bad behavior.

• 10:47Speaker 00

  Similar fundamental understanding, right? If you're running inference on the car, you still have to like figure out which of the multiple hypotheses of these future worlds you want to take action to steer it towards.

• 10:58Speaker 02

  Yeah. Okay, and these are all downstream of the same foundation model?

• 11:03Speaker 00

  That's right. So start with a foundation model.

• 11:05Speaker 02

  Yep.

• 11:05Speaker 00

  You know, then you specialize and fine tune, still off-board model, those are the teachers. And then you distill, each one of the teachers kind of distill, trains its own student.

• 11:16Speaker 02

  Yes.

• 11:17Speaker 00

  And the driver, the simulator, the critic.

• 11:19Speaker 02

  Yes. You started working on self-driving 20 years ago.

• 11:22Speaker 00

  Yeah.

• 11:23Speaker 02

  As you think about the tech evolution, is this just a scaling law story where we had to be able to throw enough compute at us? Were there architectural approaches we needed to wait to have be invented? Was it just a story of we needed 20 years of going down the wrong cul-de-sacs before we eventually arrived at the right approach? You know, could you, knowing what you know now, could you have a successful Waymo in market in 2015? Or was there some enabling technology?

• 11:55Speaker 00

  No. Technology breakthroughs that happened over the years were critically important, primarily in AI, but also in other areas, like compute. You know, they have a compute that you need.

• 12:07Speaker 02

  Yep.

• 12:08Speaker 00

  I wouldn't characterize it as like going, you know, a thousand different dead ends and then going to retract and then finding like the one right bat. I would characterize it as iterative learning and evolution. And then, you know, transformers came around. But transformers, for example, are very general architecture, right? Powers of L M's, powers, you know, our models. But how you apply them to that space, I think this is where.

• 12:33Speaker 02

  It didn't just fall out of transformers.

• 12:34Speaker 00

  Exactly, right. And of course, you know, people like to talk about architectures, but architecture is important. But really, a lot of it comes down primarily to your metrics, to your evaluation mechanisms, to all of the training recipes and of course, you know, data.

• 12:50Speaker 02

  Yes. L M's are good as text or tokens specifically. And obviously perform best at domains that have some kind of single corpus of text they can work on, like coding, where it's very helpful that everything was just kind of textual already. And part of the success has been creating textual representations for domains such that we can then, you know, put L M's against them. Can you describe how you encode the worlds that you're seeing? I mean, are you just building a 3D map, like a 3D bitmap essentially? Or?

• 13:27Speaker 00

  So this is where I think we can get a bit into the

• 13:35Speaker 00

  question of what is the interface between the encoder and the decoder parts? And I think that touches also on the thing you flagged earlier, where people like to debate end to end or not end to end. So the way

• 13:52Speaker 00

  let's talk a little bit about end to end and then get back to like, what is the interface between those two? Right? So when you say end to end, what do we mean? We mean that it is some large ML model. Typically, you don't build them monolithically. You have different parts and different subgroups. But what's important is that you can propagate back props, the gradient and the loss function all through the different layers. So they can, every layer, you can learn the weights and the representations that matter for the final task. You don't force it through some narrow funnel between, let's say, the encoder and the decoder.

• 14:29Speaker 02

  Yeah, I think of a simple view of end to end being, you know, pixels go in and car actions come out, which is maybe a bit of an oversimplification, but yeah.

• 14:37Speaker 00

  Yeah, that's exactly right. And if, like, this is kind of the basic vanilla version of it, right? There, if you think about the, you know, what will it take to build the driver that's capable of fully autonomous operation? You think about this entire ecosystem of the driver, the simulator, the critic? If that's all you do, like pixels in, trajectories out, it becomes very difficult to do all of those three and achieve the high level of safety and performance that we require. And it becomes very difficult to kind of do it at scale. And however, if, you know, that's, it's kind of a very easy way to get started, right?

• 15:21Speaker 00

  You collect some data, kind of like, and allows you to the LLM world, right? The easiest thing you can do is have, you know, pick a model. The easiest way to get started nowadays would be just take a VLM.

• 15:36Speaker 00

  It already has a language-aligned camera encoder.

• 15:40Speaker 02

  Yep.

• 15:40Speaker 00

  And then it has a decoder that will can predict, generate text.

• 15:46Speaker 00

  And you can fine tune it and say, hey, instead of text, generate trajectories. You know, very, very doable. In fact, we, you know, a while ago, we published a paper.

• 15:54Speaker 02

  Yes.

• 15:54Speaker 00

  Called AMA that did exactly that.

• 15:56Speaker 02

  Yes.

• 15:56Speaker 00

  And it will actually, I mean, in the nominal case, drive pretty darn well.

• 16:04Speaker 02

  Mm hmm.

• 16:04Speaker 00

  Which is mind-blowingly impressive.

• 16:06Speaker 02

  That is very funny, yeah.

• 16:07Speaker 00

  And I mean, there's some intuition.

• 16:09Speaker 02

  You're saying you can take an off-the-shelf model, which has nothing to do with driving to start with.

• 16:13Speaker 00

  That's right.

• 16:14Speaker 02

  And you'll get these good results.

• 16:15Speaker 00

  That's right, yeah. You get it. In the nominal case.

• 16:16Speaker 02

  Yeah.

• 16:16Speaker 00

  I just want to be clear, it's orders of magnitude away from what you need.

• 16:21Speaker 02

  Yeah, you should not try it on the street, but it works.

• 16:23Speaker 00

  For example.

• 16:23Speaker 02

  It's like a talking horse. It's impressive that it's talking, you know?

• 16:25Speaker 00

  Exactly, exactly. And you can actually, the product that you wanted to build was maybe a driver-assist system, not a fully autonomous system.

• 16:34Speaker 02

  Yep.

• 16:35Speaker 00

  Then maybe that's all you need to do.

• 16:36Speaker 02

  Yep.

• 16:37Speaker 00

  And then for that, you don't need all this other machinery of the simulator and the critic. So that this, because the number of nines is drastically. But there is, this is interesting because there, you know, there is some intuition behind, you know, why that works. If you think about the hard parts of driving, it's, you know, not unlike, you know, having a conversation.

• 16:58Speaker 02

  Mm hmm.

• 16:59Speaker 00

  Except if in the LLM world, right, having, you know, you're modeling

• 17:05Speaker 00

  language, or maybe modeling a dialogue in the space of sentences and words. What makes driving hard is also this kind of multi-agent, social, interactive part of it. And if I do something that's going to affect you, it's going to affect somebody else. And the history matters. It's not local and just geometric. Context matters, semantics matters. So, but it's in a different, you know, it's not in the language of words. It's in the language of kind of body language, if you will, right? How? So, and we see that empirically validated if you do this approach. Okay, so then let's say we build this thing. Just cameras, camera encoder, pixels go in, trajectory go out. The quality is sufficient to drive.

• 17:49Speaker 00

  In the normal case, it's not sufficient to deal with the long tail of the edge cases and hit the high bar of superhuman safety that we require. So then you start asking the question, what else do you need? Yes. And if all you did was kind of observing how other people drive when you trained the system, maybe observing just passively how people drive and how they interact, maybe also driving the car yourself and then using imitative learning to train it. Mind that that's not enough. You have to do something in closed loop. You have to do things like RLFT, which is also parallel to what we see. RLFT?

• 18:33Speaker 00

  RLFT, Reinforcement Learning Based Fine-Tuning. Okay, yes. Similar to the reinforcement learning with human feedback in the LLM world. You want to do maybe closed loop, proper closed loop driving, where you explore all kinds of different situations and then you give it a reward signal to keep it in distribution. For that, then you need a realistic simulator. Right? You also, if you want to have a good RL system, you need to have an opinion for the reward function. This is where the critic comes in, right? If you have a purely end-to-end system, let's look at the simulator. Now, what do you do? You're then constrained to just go from pixels to trajectory, right?

• 19:18Speaker 00

  That's all you can run the system on, right? And it's a very high-dimensional space. It's a hard problem to generate everything. But even if you solve that, it just becomes incredibly inefficient to run it in the full way of pixels to trajectories and simulation for training or for evaluation. So this is when intermediate representations come in. There are some intermediate representations in the world, in this task, in the physical world, we know are correct. They're not sufficient, but they're not generality limiting, right? There's an object here, there's a concept of a road, there are signs, there are speed limits. So this is where augmenting that learned representation,

• 20:02Speaker 00

  those learned embeddings from the encoder-decoder, with that more You know, structured representation.

• 20:09Speaker 02

  Yes.

• 20:09Speaker 00

  Is what we do. And we find that this kind of gives us additional knobs to simulate.

• 20:16Speaker 02

  Yes.

• 20:16Speaker 00

  You know, in that space, just pixels to.

• 20:19Speaker 02

  Yep.

• 20:19Speaker 00

  Directories. It allows us to have additional safety validation.

• 20:24Speaker 02

  Yep.

• 20:24Speaker 00

  Layers in real time. And it also allows us, you know, it gives us additional mechanisms to specify the reward function, you know, for evaluation, the critic or, you know, for training. So this is again, like we've gone kind of full circle of is it an, yes, it is.

• 20:38Speaker 02

  Yes.

• 20:39Speaker 00

  But if you want to do it at scale for full autonomy, it's augmented with all of this other stuff.

• 20:44Speaker 02

  That's very interesting on the simulating point. It's just very hard to simulate for an end to end model because it's easier to deal in end to end, or it's easier to deal in intermediate representations rather than coming up with the pixel perfect view of the world.

• 20:57Speaker 00

  You need both.

• 20:58Speaker 02

  Yeah.

• 20:58Speaker 00

  So, you know, having an antenna architecture that's augmented.

• 21:02Speaker 02

  Yes.

• 21:02Speaker 00

  With that structure allows you to kind of play in both of those worlds.

• 21:05Speaker 02

  Yeah. What are you looking to do as a self-driving car? I mean, it sounds funny, but I think people maybe don't realize that there are many different things that you're looking to solve for, where you're looking to get the person to their destination, you're looking to get them there reasonably promptly. But also drive quite smoothly and also have many lines of safety. And also not annoy other drivers and get honked at and you know, and, and, and. So what are some of the reward functions or kind of things you're optimizing for that maybe are not obvious to people?

• 21:37Speaker 00

  So safety is the primary focus, right? But of course, we also want to be a smooth driver so that for both people in the car and other actors.

• 21:51Speaker 02

  Yep.

• 21:51Speaker 00

  And also want to be, you know, a predictable well-behavior so that it can nicely.

• 21:56Speaker 02

  Yes.

• 21:57Speaker 00

  Fit into the whole social ecosystem of our, our roadways.

• 22:01Speaker 02

  It seems like one of the issues that.

• 22:07Speaker 02

  Has quickly emerged with self-driving is the fact that people can't have nice things or not everyone is nice to the robots. And so whether you're driving through a dodgy area or getting blocked, or maybe I'm not going to drop you off here. Maybe I'm going to go around the block and drop you somewhere better. But all of these, as you say, kind of other human issues. How do you go about solving this?

• 22:31Speaker 00

  A lot of the ones that you mentioned are just things that we need to work on. And understanding, honestly, you said that if we're not dropping you off where exactly where you want it to be dropped off, or we don't give you a good interface to tell us, that's on us. I just make it better.

• 22:53Speaker 02

  It feels like the drop-off is actually a pretty nuanced part of the self-driving journey, like the highway stuff and the 35 mile an hour roads like that is all nailed, but there's just like a lot of nuance in the drop-off experience.

• 23:06Speaker 00

  I'd say they're all hard. You picked freeways and you picked drop-ups for different reasons. For drop-ups, you're absolutely right. There are a few things that are maybe not obvious. You just think about this problem, but it's understanding where you want to go and making it as convenient as possible for you. And pickups from drop, it's not exactly symmetrical. But then also understanding the context of the situation where you, where do you stop? You don't want to block a driveway, you don't want to double park, although in some cases where if it's a quick one, maybe it's okay. So there's a lot of nuance that goes into doing that well, so that it's a smooth, less frictionless experience for the rider, as well as other folks.

• 23:49Speaker 00

  Freeways, for most of the time, not much happens. They're very well structured because we designed them that way. But there is still that long tail of really complicated.

• 24:09Speaker 00

  Stuff that happens

• 24:10Speaker 02

  Yes.

• 24:10Speaker 00

  Where the consequences of a bad event

• 24:16Speaker 02

  Much more

• 24:17Speaker 00

  Are much more severe, right? And the speed is much higher, everything is quadratic in speed. But we see a lot of stuff. Imagine grills falling off a freeway. Imagine people getting into accidents and kind of spinning out of control.

• 24:29Speaker 02

  You see one of those flatbed trucks, but just like a bunch of stuff piled

• 24:33Speaker 00

  Yeah.

• 24:33Speaker 02

  In it and you're driving behind it. I don't know, I always find it very nerve wracking. Looks bad.

• 24:37Speaker 00

  I know, yeah.

• 24:38Speaker 02

  Yeah.

• 24:38Speaker 00

  And we like, we've seen them, you know, leave a trail.

• 24:43Speaker 02

  Yes, yes. Yeah, okay. So it's a different set of problems. But it feels, I feel like the general sentiment with Waymo is that the

• 24:53Speaker 02

  driving has mostly now been solved by you guys. And it's kind of a question of scaling up and maybe some super long tail stuff, really snowy conditions. Like, is that your sense internally? Or is there actually much more nuance to it than that?

• 25:04Speaker 00

  I would say the, yeah, it's not like, you know, we're done with engineering.

• 25:08Speaker 02

  Yeah.

• 25:08Speaker 00

  I would say that we've clearly moved past the stage of scientific research and kind of deep core technology development to this new phase of accelerated global scaling and deployment.

• 25:25Speaker 02

  Yes.

• 25:26Speaker 00

  So, you know, we still have work to do, right? But I don't see today any limitations or any gaps in the core technology.

• 25:34Speaker 02

  The driving is good enough now.

• 25:36Speaker 00

  Well, the core technology, I think, is good enough that I can't, you know, think of any, you know, aspect of driving that is not supported by the fundamental technology. Now, that said, there is, you know, a lot of work to do in specialization and invalidation before, you know, we can deploy responsibly, right? We're not driving everywhere in the world. You know, we are planning to start operating in London and in Tokyo this year. And, you know, are we, just do have a driver that, you know, you were using today in San Francisco, that we can just plop down in London and go? No, right? But what we're seeing is incredibly encouraging from the perspective of like, is the core technology there?

• 26:20Speaker 02

  Yes.

• 26:20Speaker 00

  So now it's a matter of, you know, collecting the data, doing some specialization and validation. And you can, the signs are different, you know, in both of those places, people drive on the other side of the road. But, you know, that's actually not that hard for computers, right? And core technology generalizes really well, but you still work that you have to do.

• 26:38Speaker 02

  Generalizes least well.

• 26:40Speaker 00

  Increasingly, we're finding, especially now that we're able to kind of hook the Waymo AI to the AI in the digital world, the VLMs and kind of inherit the general world knowledge from VLMs. We're seeing really strong results from like zero-shot or few-shot learning because of that general knowledge that we bring in. But there are a few things like, say, cold weather, cold winter weather, where it affects the entire stack. Right? So it's not just, you know, the AI. We actually have to, you know, you need the hardware, you need to have the proper cleaning solution, you know, heating elements in it. And then you think about things that are completely solvable by computers, like motion control and slippery surfaces.

• 27:24Speaker 00

  Right? So that takes a bunch of work. You don't get that for free from just, you know, pulling in some, you know, VLM decoder.

• 27:32Speaker 02

  Was it the case, I mean, my impression, not knowing anything, is that in the early days, there was maybe a lot of San Francisco specific work or Phoenix specific work in the early markets, whether it be mapping or something else. And that you guys seem to either have solved that in generalizing it or just scaled up your ability to do the city-specific work. What enabled the kind of the rapid city expansion?

• 28:04Speaker 00

  We usually think about it, you know, the capability of the Waymo driver as well as deployment, not primarily indirectly in that space of, you know, cities or ZIP codes. I think about the operating domain, right? And then that's just the

• 28:18Speaker 02

  Freeways and cold weather.

• 28:19Speaker 00

  Freeways, cold weather, snow, rain, fog, density, et cetera, et cetera. And then that, that's what we are building, that's where we're evaluating. And then that maps to a city, like a particular city, be within the operating domain or outside of it. Right? So what, where

• 28:37Speaker 00

  You know, if we provide history a little bit, our initial deployment in where we started offering a fully autonomous commercial service for the first time was in 2020 in Chandler, Arizona. So, and that was on what we called the fourth generation of the Waymo driver. This was the, if you remember, the Pacifica minivans with different hardware, different software. There, you know, we were super focused on doing the whole thing end to end. Learn how to build the driver, evaluate it, deploy regularly, operate it end to end, 24/7 with customers, learn from the customers. And then we're very focused on that operating

• 29:19Speaker 00

  domain of mostly Chandler, which is a medium low complexity one. Then when we made the jump to the fifth generation of our system, this was on my basis today, we really wanted to take a huge bite out of that operating domain. And we collected data all over the United States, all different states, different cities. When we chose to deploy in the hardest parts of San Francisco, hardest parts of Phoenix, we made a big jump on the hardware side and most importantly, on the software, the AI side. And I would say that was the big discontinuous jump. And that's what you're seeing now after we've scaled up and iterated

• 30:01Speaker 00

  on all of the aspects of building and deploying driver. This is now why you're seeing us go in parallel and scaling in the US and.

• 30:11Speaker 02

  So driver version five was just a much more generalizable stack than version four? And was it about, was it just that it had been trained on a much wider dataset?

• 30:24Speaker 00

  It was when we made this big bet on AI. I think there was a lot more little AI models and ML models in the fourth generation. We made a much bigger bet and jump to kind of AI as the backbone for the fifth generation.

• 30:38Speaker 02

  AI is the backbone.

• 30:42Speaker 02

  As the core engine, as in you're saying that Gen 4 had lots of small little AI subsystems.

• 30:48Speaker 00

  Yeah. Yeah. And that's been, so we made that jump and we've been iterating and improving the model since then.

• 30:56Speaker 02

  Can we talk about hardware a second? So lots of hardware questions, but one is maybe everyone in this space has a very charismatic demo of a vehicle that is custom made for self-driving. And so, you know, it's often the van with the, you know, no steering wheel, seats facing in both directions, you know, you guys have one, Tesla has the steering wheelless Cybercab, you know, Cruise had the Cruise Origin. And yet, we're still driving in Jaguars that have a steering wheel in the front and are pretty similar to consumer cars.

• 31:40Speaker 02

  And it's interesting to me because, you know, if we were talking about this 10 years ago, we might say, well, yeah, developing a custom car, like, that's relatively straightforward. We know how to put a bunch of sensors on a new car, but the software will take a long time. And what's interesting is we've made huge progress in the software, but interestingly, the cars are still derivatives of, you know, cars that people are driving. And so I'm curious why you just think the custom hardware has not happened as of 2026? It's obviously, it's a small improvement compared to, you know, Waymo is the big improvement, but it's just interesting that it still has to happen.

• 32:14Speaker 00

  Well, let's say our sixth generation of the vehicle and the driver is our version of that.

• 32:22Speaker 02

  Oh, no, I know it is.

• 32:22Speaker 00

  It is the oh hi, one from, right? So that is, you know, it still has the, you know, we can talk about, you know, whether you want to have the seats pointed backwards or not, I actually, you know, think it's, it looks nice in a demo, but practically speaking, maybe not the way to go. But that is, it is a custom-designed vehicle and it is, we put a lot of thought into, you know, moving away from a car that's designed around the driver to a car that's designed around passenger. And it's much more spacious, like, but it's happening. It's, you know, we're not, it's not open to the. The public yet, but you know, I took a ride in it the other day fully autonomously. And that's coming this year.

• 33:02Speaker 02

  Yes. How much better is it as a passenger experience?

• 33:05Speaker 00

  You'll tell me, once you give it a try, I love it.

• 33:07Speaker 02

  Okay.

• 33:07Speaker 00

  It's so it's, yeah, it's all about the space.

• 33:11Speaker 02

  Yeah.

• 33:12Speaker 00

  And the convenience of, you know, ingress and egress and the screens and the interface of the passenger. So we put a lot of thought into every aspect of it.

• 33:23Speaker 02

  Yes.

• 33:23Speaker 00

  It has sliding doors. It's very easy to get in. It has a flat floor. It is, yeah, if you sit in the back, you can like fully stretch out and there's so much space there. And it it looks, you know, from the outside, you know, it looks fairly big.

• 33:36Speaker 02

  Yes.

• 33:36Speaker 00

  Right. But the actual footprint of that is barely, barely, barely larger than the i-Pace. So it's kind of amazing that, you know, you walk in just it feels like you're in a living room.

• 33:47Speaker 02

  Yes. I guess my question is just, you know, Waymo does, you know, 25 million rides a year, run rate ish. With the Jaguar i-Pace and it's interesting that so much scaling has happened with self-driving so far. On the old, you know, retrofit, uh, maybe that's to be expected.

• 34:08Speaker 00

  Well, um.

• 34:12Speaker 00

  It matches the high, I don't think, you know, as a given, you're right. I, uh, but if you think about the value proposition, right? Of course, there is the safety of it.

• 34:22Speaker 02

  Yep.

• 34:23Speaker 00

  Not to worry about. Uh, there's also the, um, privacy.

• 34:28Speaker 02

  Yep.

• 34:28Speaker 00

  Being in the car by yourself, maybe, you know, with other folks, but not having to share that space with another human. Right. Maybe.

• 34:34Speaker 02

  No, Waymo's a great product. Yeah.

• 34:35Speaker 00

  But I guess this is why we're seeing such, uh, you know, consistency in car, you know, drives well.

• 34:40Speaker 02

  Yes.

• 34:40Speaker 00

  You know, very predictable and, um, you know, you can go beyond that, right? And you specialize even more to make the experience even more magical.

• 34:49Speaker 02

  Yes.

• 34:49Speaker 00

  Uh, around the rider, but I guess it's, you know, it would have been disappointing if, you know, without the specialized car. And I think I would have been surprised if we leveled off, you know, at some other much lower level of customer. Because, yeah, a car seems like, you know, more of an optimization improvement, but the core of the. Value proposition comes from those other factors.

• 35:08Speaker 02

  Yes, yes. I guess to just take risk on one thing at a time will start by, you know, doing the software layer and then we'll build a specialized car or something like that.

• 35:16Speaker 00

  That's right. Yeah, it's also, I mean, as you said, it's a big investment. So you have to like, you de-risk the fundamentals.

• 35:22Speaker 02

  Yes.

• 35:23Speaker 00

  And, you know, throughout our history, we were very focused on setting the most, you know, the biggest goal for the company to de-risk the most important questions. Right. We talked about, you know, the third generation where, you know, we wanted to deploy something and go end to end. We talked about what was the goal with the fourth generation and then, oh, sorry, the fifth generation. And then there's the sixth generation, right? So it was the sixth generation where it's made sense to go out and spend all this, you know, effort into the custom.

• 35:51Speaker 02

  And sixth generation is both a custom vehicle. Is it also a new generation of the, the driving stack?

• 35:58Speaker 00

  Yeah, it is the new hardware.

• 36:01Speaker 02

  Yep.

• 36:03Speaker 00

  The sensors, you know, the hardware, the software hardware they're putting on the, the Ohi vehicle is the sixth generation. It is very different from the fifth generation. It is simpler. It is more capable. It is much lower cost. It's a thing of a fraction of the cost is, you know, comparable to what you would get like a fancy AD system nowadays, the driver assist system. The software is pretty much the same. So that's another, so when we talk about generalizability of the Waymo driver.

• 36:30Speaker 02

  Yes, yes.

• 36:30Speaker 00

  You know, we talk about weather conditions, we talk about cities, but the, it also generalizes well to different vehicle platforms and different sensor configurations.

• 36:39Speaker 02

  Okay, so Gen six is a new vehicle and a new sensor stack, but a similar, it's almost a tick tock cycle happening here. It's a similar software.

• 36:47Speaker 00

  That's right. That's right. And then we're going to put the sixth generation Waymo driver on other vehicle platforms like the Hyundai Ioniq that's coming, you know, later in the year.

• 36:59Speaker 02

  What is different about the sixth generation hardware stack and how did you make it cheaper?

• 37:04Speaker 00

  So it still has the same three sensing modalities, but we've made significant optimizations in all three. So unification, simplification, and there's just, yeah, the kind of just writing the.

• 37:21Speaker 02

  Yeah, is it a classic case of manufacturing scale?

• 37:24Speaker 00

  Well, scale hasn't fully come in place, but all of those, if you think about the supply chains, the industries. Cameras is pretty mature.

• 37:35Speaker 00

  Radars, way many years ago, used to be bulky, complex, very expensive. You know, when we were putting them in planes. But then we start putting them on cars. Now you can get a decent automotive radar for tens of dollars. There is a variant of the automotive radar.

• 37:56Speaker 00

  It's called an imaging radar. It gives you a richer something. So that is also, you know, has come down and caused drastically. But it's a little bit behind your standard automotive radars. LiDARs are following the same very predictable, very well-known trend. So we're writing that and we're also learning from the previous generation to just make improvements and simplifications and optimizations.

• 38:16Speaker 02

  It's a very silly question. What are LiDARs versus radar as in a self-driving context?

• 38:23Speaker 00

  LiDAR.

• 38:23Speaker 02

  Are they complementary?

• 38:24Speaker 00

  They're very complementary.

• 38:27Speaker 00

  You know, it's all blasting.

• 38:33Speaker 02

  Effectively.

• 38:35Speaker 00

  Effectively, like blasting photos out there. And then they bounce off of something, they come back. You know, you measure what comes back. The frequencies are very different. So laser gives you, it's very, very high resolution. So you can think of it as like a laser beam that goes out, spins around. It shoots out millions of these laser pulses per second. And then each one comes back and you can, you know, you're kind of sampling the 3D structure of the world with very high resolution.

• 39:05Speaker 02

  So LiDAR for very fine-grained mapping.

• 39:07Speaker 00

  That's right. Radar has much lower resolution, but because of the physics of it, it can, it degrades much better in adverse weather conditions. So fog, snow, heavy rain.

• 39:24Speaker 02

  So it's not going to be acceded by particles between it and the target.

• 39:27Speaker 00

  So imagine driving in. Super dense fog.

• 39:32Speaker 02

  Yes.

• 39:32Speaker 00

  We're close to San Francisco, so you probably don't have to think that hard. It can be really hard to see, so cameras degrade.

• 39:40Speaker 02

  Yes.

• 39:40Speaker 00

  Laser, depending on the size of the particulates, can degrade better or worse than camera. Radar is not well affected, so you can imagine driving on a freeway, then radar will give you really good returns for cars that are absolutely invisible in the camera space.

• 39:57Speaker 02

  That's interesting. So does that mean there are some environments where you'll be relying significantly more on radar? But the performance is good enough?

• 40:05Speaker 00

  Well, it's a combination of the sensors, right? So we rely on, each one is noisy, right? How the noise characteristic show up in different environments is different, but it is, I mean, it's not like we switch from one to another. It's not like we estimate what's happening with the world through cameras and through radars and through radar, and then we compare. No, there's an encoder for camera, there's an encoder for lighter, and they all go into the system that gives you jointly the best view of what's happening in the world. So if it's a nice, bright, sunny day, cameras are very valuable. If it's pitch dark, or you have sun in your face, or you're blinded by the headlights

• 40:49Speaker 00

  from an incoming car, then camera will degrade. There's still some noisy signal, but it will degrade.

• 40:55Speaker 02

  Yes.

• 40:55Speaker 00

  And radar, lighter is completely unaffected, right?

• 40:59Speaker 02

  Are there technical problems that are your white whale, or you're still chasing, or you are particularly interested in solving? Even if they're kind of niche for the, we really want to have, driving when it's actually snowing nailed, or steep hills in San Francisco, or are there problems you've been very interested in historically, or still are?

• 41:21Speaker 00

  I'm super excited right now about the accelerating global expansion. More cities in the United States and going internationally. So being, I don't understand, I'm not answering your question about the technology, I'll come back to that, but really that's the thing that I'm today most excited about, just going to get into a place where any major...

• 41:47Speaker 00

  You know, metropolitan area, you can fly into the airport and then take a Waymo and go anywhere you want to go. Like, that is

• 41:53Speaker 02

  Yes, yes.

• 41:54Speaker 00

  insanely exciting to me right now. So then, you know, technically, uh, what, uh, I'm most excited about is, uh, all of the rapid progress in AI.

• 42:07Speaker 02

  Mm hmm.

• 42:08Speaker 00

  And, uh, the world models, the foundational model work. And it is just such a massive boost to how much we can simplify the system.

• 42:21Speaker 02

  Yep.

• 42:21Speaker 00

  How much we can bring down the cost and how we can scale globally.

• 42:24Speaker 02

  Mm hmm.

• 42:25Speaker 00

  And there's some magic that happens that I don't think I would have anticipated, you know, a few years ago.

• 42:31Speaker 02

  Yes.

• 42:31Speaker 00

  So that I find from the technical perspective, just insanely thrilling.

• 42:36Speaker 02

  Yes. When you talk about kind of the progress in AI, what are the most fun parts of it for you these days?

• 42:42Speaker 00

  I think it's seeing the capability and the scaling laws from this approach of starting, you know, with that cornerstone of the foundational model and then specializing to T-shirts and then, you know, distilling.

• 43:00Speaker 02

  Mm hmm.

• 43:00Speaker 00

  It just, uh, you get such big wins, uh, in, in performance across the board. I just, you know, you, you know, invest something into, you know, the architecture or get better data or training recipe.

• 43:13Speaker 02

  Yes.

• 43:14Speaker 00

  And then, yeah, uh, you, you invest in that early stage.

• 43:17Speaker 02

  Yeah.

• 43:17Speaker 00

  And then it just has massive amplification ripple effects. So that, uh, is in some ways is kind of magical. And then you, I guess, then you see it on the car. And I've had some moments where, you know, a car does something and you look at a log. And I've been surprised, like it does things that I didn't think it was capable of doing.

• 43:42Speaker 02

  Hmm.

• 43:42Speaker 00

  Right, so it's, it's that, uh, you know, it, it, it's

• 43:47Speaker 02

  When you see emergent behavior, that's kind of

• 43:49Speaker 00

  That's right, that's right.

• 43:50Speaker 02

  A proud moment.

• 43:50Speaker 00

  One example, yeah, you know, it's, you know, when you build a system and then, you know, you think you understand, you know, how it works and you understand fully, you know, the limits of its capability and performance. And then it does something. Kind of almost magical. Yes. It's it's exhilarating. Yes. So one example I can give you, I think I've shared some videos of that publicly in some talks, was this example where the situation, it happened in San Francisco. Our, you know, fairly benign situation, we're at an intersection, our light is red. There's cross traffic, a bus goes by and, you know, it stops partially blocking.

• 44:32Speaker 00

  I mean, our light light turns green, so we start to go, we're nudging around the bus. And then you see a pedestrian being detected on the other side of the bus. Yeah. Right. And then your car responds appropriately, it slows down, goes a little bit wider. And, you know, then pedestrian actually emerges from the bus and, you know, we go on our own way. So the first time I looked at that log, like, what's going on here? Like, I I know we have pretty darn good sensors and the software is very capable. Like, we don't see through. Yeah, yeah, yeah. Right? Okay, that's not how cameras or lighters and radars work, right? I saw the pedestrian through the bus. It threw, I saw the pedestrian on the other side of the bus. Yeah, yeah.

• 45:13Speaker 00

  And it's not like, you know, you look at the windows and like, okay, you know, radar shouldn't, this massive metal box. Yeah. Yeah, you know, look at the sensor data. Yes. And like, it shouldn't, radar shouldn't be able to go through it, right? Um, you know, the camera, like, you can't see in the camera because, you know, there's reflections and there's people on the bus. So it's not like you can see through the windows. Right. So like, what is going on? Maybe it's, you know, noise or some coincidence. Uh, and I, you know, first time I saw it, I like, I couldn't actually believe it. It's like, no, no, there's something, this doesn't smell right. So what actually turned out was happening is that our peripheral lighters and bombs under the bus. And there was just a little bit of very, very noisy reflection of the movement of the person's feet.

• 45:54Speaker 00

  That was enough for the AI models, they, hey, I likely there's a pedestrian there. And I'm gonna, you know, uh, you know, I detected a sight. Yeah, yeah, yeah. And moreover, there's enough data there to predict what they're going to do. Yes. It's kind of like blew my mind.

• 46:11Speaker 02

  Is this the perfect example to explain what we were talking about earlier? The value of one fusion across a sensor suite, but then secondly, building, I mean, relatedly, building an intermediate representation of what's going on, where if you're just dealing with pixels, I mean, the person behind the bus does not exist in pixel space. And so you need to have some representation of the world that exists to be able to reason about the person behind the bus.

• 46:43Speaker 00

  I think it's an example where giving it kind of,

• 46:49Speaker 00

  using that intermediate representation to boost the level of performance of all parts of the model is what's happening here. Just imagine solving this problem with a black box, you know, purely open loop, imitative system. Be, is it, you know, impossible? No. In practice, what would it take to achieve that level of performance?

• 47:14Speaker 02

  Yes.

• 47:14Speaker 00

  Very, very difficult.

• 47:15Speaker 02

  What metrics can you share on just where the business is at today in terms of rides, revenues, cars on the roads?

• 47:25Speaker 00

  We have about 3,000 cars on the roads. We're doing about half a million rides per week. That translates to about over 4 million fully autonomous miles per week. We are operating in a fully autonomous mode in 11 cities in the US. 10 of those, we have riders, riders.

• 47:54Speaker 02

  What's the ghost city?

• 47:55Speaker 00

  The ghost city is Nashville. We just started there. So we just opened it up to riders in four new cities in one day. That was one of those, you know, little but super exciting moments where I thought back to the history, like, how long did it take us from the first time we started fully autonomous rider only operation to the first time we had external riders in four cities? It's about eight years. And then just, you know, the other week, we just launched four in one day.

• 48:28Speaker 02

  Yes, yes. It seems now clear that in Fifteen years, most miles that are driven will be autonomous. Like, there'll be some burning period and, you know, there's lots of old cars in the road. Like, I think it'll actually take a little while. And some of that will be by level four, level five systems, expanding in new cities. And that expansion continuing. Some of it will be, you know, you referenced existing driver assist systems and kind of getting up to, you know, level two and level three and existing systems in, across current car brands, getting more and more capable. What do you think that working your way up from the lower levels versus working your way,

• 49:12Speaker 02

  expanding from existing products like Waymo? What'll that convergence look like? Because we're going to eat it from both sides.

• 49:21Speaker 00

  I don't believe we will. And I actually think this.

• 49:25Speaker 02

  That's a great answer.

• 49:28Speaker 00

  Cars will get smarter. There's going to be advances in driver assist systems. And if there is, at the same time, from level four autonomy, you know, there is simplification. And the sensors of today are not going to be the sensors of tomorrow. So they'll be much more integrated, they'll be simpler, there'll be much lower costs. So from that perspective, they're gonna, you know, there is a path of convergence. And there's also, you know, a path of convergence from, you know, the product lines. Right-hailing and what, you know, you can take, you know, a ride through the Waymo app today. You know, eventually they'll be on your personal car. So that I see. And talk about the technology and I see it just as fundamentally two different problems.

• 50:11Speaker 00

  There's driver assist systems and then there is full autonomy. And I think it's deceptive to think of them as kind of incremental, you know, on one spectrum of complexity.

• 50:24Speaker 02

  Okay, but you think one cannot work one's way up from driver assist systems to full self-driving? You think you have to start building a full self-driving system?

• 50:34Speaker 00

  I think you have to tackle, if I think about the hardest parts of.

• 50:42Speaker 00

  Building a fully autonomous rider-only system, they are very different from what you do for a driver assist system.

• 50:51Speaker 02

  Yep.

• 50:51Speaker 00

  And of course, some work in this space helps you, right? So I don't want to say you can't make the jump, but it is a qualitative jump.

• 50:58Speaker 02

  Yes. When can I buy a Waymo so that I don't need to wait for it when I want to go? I can just like when I'm ready, I can walk out the door and it's there.

• 51:07Speaker 00

  I'm not going to give you a date today, but you're not the first person to bring this up as a

• 51:11Speaker 02

  That's my product request.

• 51:12Speaker 00

  The product request. Yeah. Do we know it?

• 51:15Speaker 02

  Okay.

• 51:16Speaker 00

  I'll add it to the list.

• 51:17Speaker 02

  Yes, you know, that waiting for the cars, it's just in the garage there and you keep your stuff in there and everything. It's not the first time you've heard that request.

• 51:25Speaker 02

  So how, it seems to me operationally very intensive and very hard.

• 51:34Speaker 02

  Like a self-driving car is actually not self-driving. It takes a village. You have all of the human operator ready to step in. And you know, there was a thundering herd incident that you guys talked about in San Francisco that kind of highlighted that for people. And then there's just like keeping the cars clean and, you know, keeping everything running in that regard. And so can you describe just what the operational infrastructure that sits behind Waymo looks like?

• 52:01Speaker 00

  Sure. And I will say that we are overall, in all of those areas, on a path of increasing efficiency and automation.

• 52:15Speaker 02

  Yep.

• 52:15Speaker 00

  All right, so you know, the number of manual steps that one had to do, you know, five years ago to, you know,

• 52:25Speaker 00

  you know, launch a Waymo versus where we are today is drastically different, right? So, yeah.

• 52:33Speaker 00

  But nowadays, if you look at one of our depots, it's like a fully automatically orchestrated dance of autonomous vehicles. So the way it looks, what it looks like today is cars will automatically go on there to pick up their riders, serve their trips. If for some reason they need to come back, maybe they're low on energy, maybe somebody left a mess in the car, they will automatically come to the depot. If it is, so cleaning today is a manual process. So it'll get flagged in the car.

• 53:15Speaker 00

  We have fleet management systems, say, hey, car number 378 needs cleaning. And we'll actually, on the sensor dome, we're able to display on console, show you like a little emoji that highlights. Yeah, and there's people whose job it is to clean cars still come clean up. If that's, cleaning is not required and it's just charging, we'll automatically pull into a charging stall and we'll say, hey, I need charging. We don't yet have automated charging. In the future, you can imagine that being fully automated, right? But a person will come in and plug in a cable and the car will charge and they'll say, hey, now I'm ready to go. And it will get unplugged and the car will pull out of its parking stall and then go on its merry way.

• 54:00Speaker 02

  One of the new Porsches, I think it is, has inductive charging, like just like your iPhone where you just drive over the charging mat. I was amazed that that works at car scale, but presumably in the future they'll just be able to drive on the charging mat. Or do you think just robotic plug-in will be easier?

• 54:14Speaker 00

  We'll see, we'll see. I don't know. I think there's some questions about efficiency and how that plays into the overall cost and which one will be most cost beneficial remains to be seen, I think.

• 54:26Speaker 02

  How well behaved are the Waymo riding population in terms of not leaving a mess in the car?

• 54:34Speaker 00

  We have wonderful riders. We have the most amazing customers in the world. Generally, I would say they are very good. I think there is something about, you know, I talked about not having, you know, a person in the car is not somebody else's car. In some ways, you kind of like want to preserve the, I think generally people want to kind of preserve the nice aspects of it, right?

• 54:59Speaker 02

  It's a broken window thing where it's so clean to begin with.

• 55:02Speaker 00

  I know, yeah. It's kind of like, you know, I think that's the general trend that we see, right? And it's like, because there's not somebody else's space, you know, you're in it. It feels like it's your own. So you don't like want to mess up, you know, your own space. I think, I mean, I don't want to speculate too much on the, you know, say, college of things. However, I will say that it varies. And you can imagine, you know, a college town on a, you know, Saturday night, and yeah, that's a different distribution.

• 55:28Speaker 02

  Yes, yes. Will I be able to go Waymo at any address that has USPS service in the US? Or will there be some head-tail dynamic where Ketchikan, Alaska is just never worth it?

• 55:44Speaker 00

  Eventually it will, absolutely. Right, there's no doubt in my mind. I think it's just a matter of when? And, you know, what modality would make the most commercial sense?

• 55:55Speaker 02

  Right share versus privately owned?

• 55:57Speaker 00

  Yeah, for right, you know, it's not a technical problem. I mean, you know, technology is solved.

• 56:01Speaker 02

  Yes.

• 56:01Speaker 00

  But then, you know, if you're in the middle of nowhere and there's just not enough density of the trips, does it make sense for the right-hailing service that, you know, Waymo is running to, you know, have cars on standby?

• 56:10Speaker 02

  Yes, yes.

• 56:10Speaker 00

  Probably not, right? They can be deployed, you know, somewhere else. And you probably don't want, you know, a horribly bad ETA. And this is where, you know, a personally owned vehicle that is equipped with the Waymo driver is maybe, you know, how you will see it materialized.

• 56:25Speaker 02

  Relatedly, what will the second order effects of, say, majority autonomous traffic be? Like, it feels like a lot of things will work better where, as you say, you know, when someone merges into a lane very poorly and everyone all the way back has to, you know, slam on the brakes, that's kind of antisocial behavior. And so it feels like higher quality and more pro-social driving will just, I mean, basically reduce traffic a little bit, even for the same number of cars on the road. But presumably, there'll be other second order effects, like we'll want higher throughput traffic lights and, yeah, how else will things change?

• 56:58Speaker 00

  So the first thing I think that you mentioned is that's a huge deal. I just think about traffic jams.

• 57:09Speaker 02

  Yep.

• 57:09Speaker 00

  What's that saying, the Navy SEALs, slow is smooth and smooth is fast? That's what like traffic jams are. Like you accelerate abruptly, then you come to a stop. And sometimes you have a traffic, you're like, what happened?

• 57:22Speaker 02

  Yes.

• 57:23Speaker 00

  Well, you know, an old lady crossed the road three hours ago and we still have the standing wave.

• 57:27Speaker 02

  Yeah.

• 57:27Speaker 00

  Right? So if everybody was a kind of a smooth predictable driver and a consistent driver and you still have those new traffic jams at the time of.

• 57:38Speaker 02

  Yep.

• 57:38Speaker 00

  But then the time constant to clean it out, I think would be very different. But longer term, and yes, things like parking lots.

• 57:48Speaker 02

  Mm hmm.

• 57:48Speaker 00

  Right now, if you look at what is our most interesting.

• 57:51Speaker 02

  Yes.

• 57:52Speaker 00

  Yeah.

• 57:54Speaker 00

  Pieces of land allocated to.

• 57:56Speaker 02

  Yes.

• 57:56Speaker 00

  You know, it's parking lots, it's garages. And why is that?

• 57:59Speaker 02

  Yes.

• 58:00Speaker 00

  Well, because again, your car is just sitting there 90% of the time. Right? If more cars become fully autonomous, then there's no, you know, that, right? And like, then imagine, just imagine what you can do.

• 58:12Speaker 02

  Yes.

• 58:12Speaker 00

  With, you know, your favorite city in the world. If you don't have to spend that money, that huge fraction of it on, you know, just, just keeping your, these chunks of metal sitting around.

• 58:22Speaker 02

  Yeah, I don't think people often realize how big a deal parking minimums are for the layouts of the urban landscape. The coffee shop here where I am would like to have outdoor seating but can't because it would reclaim parking spots.

• 58:33Speaker 00

  Yeah, wouldn't it be wonderful?

• 58:34Speaker 02

  Yep. I have a few more questions, but I'm curious to talk about Google's relationship with

• 58:43Speaker 02

  self-driving, where again, it feels like right now Waymo is, aside from everything else, AI related, kind of the most exciting thing happening at Google. But it was a very long journey to get here. I mean, I feel like you could say that Google almost started working on this too early because you were saying there's been a bunch of recent enabling technologies. And so did it require Google starting when it did so early? Or could one have spun up this project in 2015, 2020? And then how did Google keep the faith?

• 59:23Speaker 02

  When it always felt like it was perennially two years away.

• 59:27Speaker 00

  Yeah, no, I know the latter part, I just have to give credit and huge kudos and gratitude to Larry and Sergey and Alphabet Leadership and their company. It is part of the culture and the DNA of the company is to have that vision and have the stamina and conviction to go the distance. So to the other part of the question, was it too early? I don't know.

• 1:00:09Speaker 00

  I think what we've been seeing, clearly, all of the breakthroughs that we've seen over the years have changed how we're building the system. But the complexity of the problem is such that you need to go through these iterative cycles, right? It's not still, and we've seen many waves of technology. There's breakthroughs in 2013, ImageNet came around. Those are, okay, like, that is the right time to start BSL Drive. And then Transformers came around, and VLMs. And all of those are super powerful and have applications in other spaces.

• 1:00:49Speaker 00

  And in the digital world, they certainly have an impact on our AI and the physical world. They're no silver bullets. They drastically reshape that early part of the curve. It's always been that the nature of this problem is very easy to get started. It's deceptively easy to get started, but it is super hard to go the full distance.

• 1:01:10Speaker 02

  In the case to me.

• 1:01:12Speaker 00

  The number of nines, right? There's the standard engineering rule of thumb that every next nine takes 10x more. So I, yeah, maybe there is a more optimal path, but I don't see there's some magical moment where the true complexity of the problem goes away and then you can just take some off-the-shelf components and you're a business. If that were the case, then I think the industry would look very different today.

• 1:01:41Speaker 02

  Last question I have, you've been promoted a lot at Google. It feels like Google really recognized your talents. Just what do you think Google does? Like Google is famously one of the very best in the world as technical talent. And say, you know, the current AI wave more broadly happening, you know, is either stuff happening at Google or generally Google alumni. But just what have you observed firsthand from how Google does this so well?

• 1:02:10Speaker 00

  Yeah, I'll say, uh, Google, you know, that, that culture of Google of not accepting the status quo, having, you know, a big vision and, uh,

• 1:02:28Speaker 00

  investing in technical talent, the people who can, you know, go the distance and realize the vision. That is part of the, the culture. I think this is what you're seeing and, uh, with the, you know, the breakthroughs in AI in the digital world and like all of the early investments in, you know, transformers and other, you know, other fundamental, uh, technologies, uh, you know, quantum computing. And, you know, I guess we're not unlike those efforts as well.

• 1:02:54Speaker 02

  It's a pleasure. Thank you.

• 1:02:55Speaker 00

  Yeah.

• 1:02:58Erik Torenberg?

  Thanks for listening to this episode of the A16Z Podcast. If you liked this episode, be sure to like, comment, subscribe, leave us a rating or review, and share it with your friends and family. For more episodes, go to YouTube, Apple Podcasts, and Spotify. Follow us on X at A16Z and subscribe to our Substack at A16Z.substack.com. Thanks again for listening and I'll see you in the next episode.

• 1:03:24Erik Torenberg?

  This information is for educational purposes only and is not a recommendation to buy, hold, or sell any investment or financial product. This podcast has been produced by a third party and may include paid promotional advertisements, other company references, and individuals unaffiliated with A16Z. Such advertisements, companies, and individuals are not endorsed by AH Capital Management LLC, A16Z, or any of its affiliates. Information is from sources deemed reliable on the date of publication, but A16Z does not guarantee its accuracy.